Octave: The first TTS powered by a language model

Octave is the first LLM for text-to-speech. Design any voice you can imagine, control emotions and style with acting instructions, and use our creator studio to generate long form content like audiobooks and voiceovers.

Octave is trained to understand and synthesize speech. This speech-language model can predict the tune, rhythm, and timbre of speech, inferring when to whisper secrets, shout triumphantly, or calmly explain a fact. In other words, Octave interprets plot twists, emotional cues, and character traits within a script or prompt, then transforms that understanding into lifelike speech, like a human actor reading a script.

Start creating: hume.ai

Learn more: www.hume.ai/blog/octave-the-first-text-to-speech-m…